The goal of this project is to develop a novel class of effective antiviral agents based on the (-)s TRSV hairpin ribozyme. Ribozymes are catalytic RNA molecules; many catalyze site-specific RNA cleavage reactions. We will determine the mechanism for substrate recognition and cleavage of the hairpin ribozyme, and use this information to rationally alter the specificity of the ribozyme by engineering sequence changes within its substrate-binding domain. The Specific Aims of this proposal are to (1) Determine the substrate selection rules for the hairpin ribozyme, (2) Manipulate the sequence specificity of the ribozyme to generate ribozymes specific for a variety of viral RNA sequences, (3) Construct and evaluate the activity of multivariant ribozymes, (4) Develop a bacteriophage system for analyzing and optimizing antiviral activity of ribozymes in vivo, and (5) Use the bacteriophage system to optimize the activity of ribozymes against foreign sequences. Ribozymes for this study are based on the self-cleaving minus strand of satellite tobacco ringspot virus ((-)sTRSV). In preliminary work, we have engineered a major change in the substrate specificity of the ribozyme, so that it selectively attacks HIV-1 pol sequences. Potential advantages of antiviral ribozymes are several: (i) Ribozymes can potentially inhibit both viral infection and the expression of viral genes in cells that are already infected, (ii) Ribozymes may be highly selective agents, (iii) Ribozymes act catalytically, so that a single molecule of ribozyme may potentially destroy many molecules of viral RNA, (iv) Ribozymes are not susceptible to pleiotropic drug resistance, and (v) ribozyme technology should be widely applicable, so that ribozymes may be rapidly developed to combat infections by newly emerging or newly recognized viruses. In addition, this work will produce a large amount of valuable information concerning the structure and function of ribozymes.